We propose to construct mice which are heteroplasmic for two different mitochondrial DNAs by microinjection of genetically marked mitochondria into fertilized eggs and subsequent reimplantation of those eggs into a suitable host mother. We will then follow the subsequent segregation of these mitochondria both in tissues of the heteroplasmic animal itself and in the normal offspring of any female animals. The results will be used to test the models of mitochondrial segregation we have proposed. These models suggest that mitochondrial DNA amplification and/or segregation during oocyte development result in the selection of minor components of the mitochondrial population which are passed on to progeny during maternal inheritance of mitochondria. If successful, these studies will be expanded to examine developmental pathways, cross-species hybridization, and the possibility of embryonic drug selection are also proposed. Additionally, if microinjection into murine oocytes is successful, we will call upon our experience with bovine oocytes and artiodactyl mitochondrial genotypes by extending these genetic manipulations into dairy cows. We will construct not only intra-specific hybrids (i.e., between cow breeds with distinct mtDNA restriction patterns), but also construct crosses between related bovid species, to measure both evolutionary distance and developmental compatibility. The methods developed here would lay a foundation for the future in vitro genetic manipulation of the mitochondrial genome.